Background Arsenic exposure is a global health concern. Several studies have focused on chronic arsenic exposure in adults; however, limited data are available regarding the potential adverse effects of prenatal exposure on fetuses and neonates. Objectives To assess which time point maternal arsenic exposure may influence the fetus during pregnancy and birth outcomes. Methods In this study, total arsenic concentrations were analyzed in urine samples collected from 130 women with singleton pregnancies (22–45 years old) in Taiwan from March to December of 2010. All fetal biometric measurements in each trimester period and birth outcomes at delivery were obtained. We applied a generalized estimating equation model and multivariate regression models to evaluate the associations between maternal urinary total arsenic (UtAs) exposure during pregnancy, fetal biometric measurements, and neonatal birth outcomes. Results We observed statistically significant correlations between maternal UtAs levels and the fetal biparietal diameter over all three trimesters (β = − 1.046 mm, p < 0.05). Multiple regression analyses showed a negative association between maternal UtAs levels and chest circumference in the first trimester (β = − 0.721 cm, p < 0.05), and second-trimester UtAs exposure was associated with decreases in birth weight (β = − 173.26 g, p < 0.01), head circumference (β = − 0.611 cm, p < 0.05), and chest circumference (β = − 0.654 cm, p < 0.05). Dose-response relationships were also observed for maternal UtAs exposure and birth outcomes. Conclusions We identified a negative relationship between maternal UtAs levels during pregnancy, fetal development, and neonatal birth outcomes. These findings should be confirmed in future studies with large sample sizes.